Geomagnetic Storm Effects at F1-Layer Heights from Incoherent Scatter Observations

c Annales Geophysicae (2003) 21: 583–596 European Geosciences Union 2003 Annales Geophysicae

Geomagnetic storm effects at F1-layer heights from incoherent scatter observations

A. V. Mikhailov1 and K. Schlegel2 1Institute of Terrestrial Magnetism, Ionosphere and Radio Wave Propagation, Troitsk, Moscow Region 142190, Russia 2Max-Planck-Institute fur¨ Aeronomie, Max-Planck-Str. 2, D-37191 Katlenburg-Lindau, Germany Received: 18 April 2002 – Revised: 24 July 2002 – Accepted: 30 August 2002

Abstract. Storm effects at F1-layer heights (160–200 km) 1995; Buonsanto, 1999; Mikhailov, 2000; Richmond, 2000; were analyzed for the first time using Millstone Hill (mid- Danilov and Lastovicka, 2001 and references therein). The latitudes) and EISCAT (auroral zone) incoherent scatter (IS) main attention is paid to the F2-layer storm effects being observations. The morphological study has shown both in- the most pronounced and impressive, while F1-layer distur- creases (positive effect) and decreases (negative effect) in bances are practically not discussed. On the one hand, this is electron concentration. Negative storm effects prevail for all due to the problems with the foF1 identification (Shchepkin seasons and show a larger magnitude than positive ones, the and Vinitzky, 1981); on the other hand, F1-layer storm ef- magnitude of the effect normally increasing with height. At fects are relatively small compared to the F2-layer ones and Millstone Hill the summer storm effects are small compared are not very important from the radio-wave propagation point to other seasons, but they are well detectable. At EISCAT of view. Theory of the F1-layer formation (Shchepkin, 1969; this summer decrease takes place only with respect to the au- Shchepkin et al., 1972; Antonova and Ivanov-Kholodny, tumnal period and the autumn/spring asymmetry in the storm 1988a, b) indicates a close relationship between F1-layer effects is well pronounced. Direct and significant correla- electron concentration and neutral composition. Therefore, tion exists between deviations in electron concentration at the observed small reaction of the F1-region to the geomag- the F1-layer heights and in the F2-layer maximum. Unlike netic disturbances followed by large perturbations in the ther- the F2-layer the F1-region demonstrates a relatively small mospheric parameters is interesting from a physical point of reaction to geomagnetic disturbances despite large pertur- view and should be explained. Recently, Buresova and Las- bations in thermospheric parameters. Aeronomic parame- tovicka (2001), and Buresova et al. (2002) have attempted to ters extracted from IS observations are used to explain the systematize the F1-layer storm effects analyzing Ne(h) pro- revealed morphology. A competition between atomic and files obtained from the ionogram reduction over some Eu- molecular ion contributions to Ne variations was found to be ropean ionosonde stations. The analysis revealed two in- the main physical mechanism controlling the F1-layer storm teresting effects: (i) independently of the sign of the F2- effect. The revealed morphology is shown to be related with layer disturbance (positive or negative), the F1-layer elec- neutral composition (O, O2,N2) seasonal and storm-time tron concentration always decreases (negative storm effect), variations. The present day understanding of the F1-region (ii) there exists a summer/winter and autumn/spring asym- formation mechanisms is sufficient to explain the observed metry in the F1-layer geomagnetic storm effects. They have storm effects. shown that the summer F1-layer (the electron concentration at 160–190 km) practically does not react to geomagnetic Key words. Atmospheric composition and structure storms, while a well detectable negative storm effect takes (thermosphere-composition and chemistry); ionosphere (ion place in winter, and also, that autumnal storm effects are chemistry and composition; ionospheric disturbances) stronger than vernal ones. These conclusions need further analysis to check whether such an F1-layer reaction is sys- tematic or the results just reflect the peculiarity of the ob- 1 Introduction servations chosen. Electron concentration both in the F2- and F1-regions is known to be strongly dependent on neu- Ionospheric disturbances related to geomagnetic storms are tral composition as the processes of photoionization and re- widely discussed in literature (see the reviews by Prolss,¨ combination are mainly the same. So, in principle, one may Correspondence to: K. Schlegel expect a synchronous (to some extent) type of NmF2 and ([email protected]) NeF1 storm variations. The results by Vinitzky et al. (1982) 584 A. V. Mikhailov and K. Schlegel: Geomagnetic storm effects at F1-layer heights

tions. Usually Millstone Hill radar provides three (some- times more) local height Ne, T e, T i and V z profiles per hour, with a 21-km height resolution. We had to use a 3– 4 h period of observation to calculate median height profiles with the standard deviations (SD) at each height. These median profiles were then smoothed by a polynomial fit up to the 5th order to be used in our analysis. EISCAT multi-pulse observations provide excellent height profiles with a 3-km height resolution in the 87–260 km height range. Due to frequent (every 2–5 min) observations, median profiles can be calculated over a 1–2 h period; therefore, for the interesting cases, two different periods within one day can be considered for analysis (24/25 October 1990). Usually the periods around noon of maximal stability in NmF2 and hmF2 variations were selected to decrease the scatter and provide reliable median Ne(h) profiles. Unlike Millstone Hill the EISCAT Ne(h) profiles are not normalized by foF2 for each particular experiment, and this may result in wrong relative deviations when disturbed and reference days are compared. Therefore, the following procedure was applied. Long-pulse observations are known to provide reliable relative Ne(h) profiles at the F2-region heights. Using the long-pulse NmF2/Ne (220 km) ratio, the multi-pulse NmF2 value was calculated assuming that relative Ne(h) height profiles are the same in the (220 km – hmF2) height range both for long-pulse and multi-pulse observations. This calculated NmF2 was then normalized by the observed foF2 value measured by the Tromsø or Kiruna ionosondes. Then the whole Ne(h) multi-pulse profile was normalized by a correspond- ing factor. The observations were grouped by 4 seasons regardless of the solar activity level. Since we analyzed only sunlit Fig. 1. The relative (disturbed/quiet) deviations δNe at 180 km ver- conditions EISCAT winter observations were not included. sus relative deviations δNmF2 for Millstone Hill and EISCAT storm The list of analyzed dates (disturbed/reference), along with observations. daily Ap and F10.7 indices, are given in Tables 1 and 2 for Millstone Hill and EISCAT, correspondingly. The ratios r = Ndist/Nref, along with absolute errors for NmF2 and seem to confirm this. With respect to the sign of the F1-layer Ne at 5 heights (160–200 km), are given in the tables. Over- storm effect, the ionogram reduction for Moscow (Zevakina all, 37 cases from Millstone Hill and 26 cases from EISCAT et al., 1971) yielded in some cases positive 1Ne at F1-layer were analyzed. The observations are seen to overlap all lev- heights during daytime hours. Therefore, the aim of the pa- els of solar activity from the deep minima in 1986, 1996 to per is to analyze the effects of geomagnetic disturbances at the maximum in 1990–1991. A strong scatter in observa- F1-layer heights (160–200 km) using Millstone Hill (middle tions for some dates (especially at Millstone Hill) results in latitudes) and EISCAT (auroral zone) incoherent scatter (IS) abnormal SD values, and such cases are marked by dashes in daytime observations, and to give a physical interpretation to Tables 1 and 2. With regard to this, it should be noted that the revealed Ne(h) variations. while some wrong (due to occasional scatter) Ne values may strongly affect the calculated SD, the median values used in our analysis are much less sensitive to such scatter and are 2 Observations more reliable compared to mean Ne values. Daily Ap index is known to be a rather poor indicator of All available Millstone Hill (zenith observations) and EIS- ionospheric disturbances. Therefore, one may find days in CAT CP-1 and CP-2 (field-aligned multi-pulse, as well as Tables 1 and 2 with relatively low Ap values, which in fact long-pulse) observations were examined to select couples were disturbed due to the preceding disturbed periods (e.g. (disturbed/reference) of days with daytime observations. We 7 May 1986, 28 September 1995 from Table 1 or 14 August have tried to choose reference days to be geomagnetically 1985, 26 March 1998 from Table 2). Due to this delay in the quiet and close in time to the analyzed disturbed ones, but occurrence of ionospheric disturbances, the reference days this was impossible in some cases due to irregular observa- may have relatively high Ap indices, but in fact they were A. V. Mikhailov and K. Schlegel: Geomagnetic storm effects at F1-layer heights 585

Table 1. List of Millstone Hill observations (disturbed/reference days), together with daily Ap and F10.7 indices. Observed rNmF2 and rNe at 160–200 km heights are given, together with absolute deviations (dashes correspond to large scatter in data when calculated standard deviations are unreliable). G-conditions means the absence of the F2-layer peak, LT=UT-5

Winter

Dates UT Ap F10.7 rNmF2 rNe(160) rNe(170) rNe(180) rNe(190) rNe(200) 15/13 Jan 1988 15:00–20:00 63/7 112.4/108.1 0.42 ± 0.13 1.08 ± 0.32 0.92 ± 0.22 0.76 ± 0.18 0.62 ± 0.14 0.52 ± 0.11 11/14 Jan 1990 18:00–21:00 18/8 169.5/165.9 1.06 ± 0.23 1.45 ± 0.92 1.34 ± 0.73 1.23 ± 0.62 1.14 ± 0.47 1.04 ± 0.46 25/23 Jan 1993 17:00–20:00 25/4 105.8/106.0 1.37 ± 0.48 1.12 ± 0.54 1.09 ± 0.47 1.06 ± 0.45 1.06 ± 0.42 1.08 ± 0.39 26/23 Jan 1993 17:00–20:00 22/4 106.5/106.0 1.15 ± 0.26 1.23 ± 0.59 1.15 ± 0.50 1.09 ± 0.53 1.06 ± 0.45 1.06 ± 0.40 30/29 Nov 1994 15:30–18:35 21/9 78.3/79.6 0.80 ± 0.20 0.82 ± 0.15 0.83 ± 0.18 0.83 ± 0.17 0.82 ± 0.15 0.80 ± 0.14 1 Dec/29 Nov 1994 15:30–18:40 18/9 79.1/79.6 1.20 ± 0.24 1.03 ± 0.18 1.03 ± 0.16 1.04 ± 0.15 1.04 ± 0.13 1.05 ± 0.16 10/9 Jan 1997 17:00–19:20 32/5 75.4/73.7 0.52 ± 0.04 0.81 ± 0.18 0.69 ± 0.13 0.60 ± 0.09 0.55 ± 0.07 0.48 ± 0.06 7/9 Nov 1997 17:00–20:00 44/11 94.4/86.4 0.58 ± 0.06 0.97 ± 0.40 0.77 ± 0.26 0.63 ± 0.17 0.56 ± 0.13 0.53 ± 0.10 11/10 Feb 1999 16:00–18:00 20/6 163.5/152.4 1.26 —- 0.90 — 0.89 — 0.87 — 0.86 — 0.86 — 14/15 Feb 2001 16:30–18:30 18/5 137.9/135.1 0.86 ± 0.05 0.45 ± 0.11 0.51 ± 0.07 0.57 ± 0.07 0.61 ± 0.06 0.65 ± 0.06 Spring

Dates UT Ap F10.7 rNmF2 rNe(160) rNe(170) rNe(180) rNe(190) rNe(200) 18/17 Mar 1990 16:30–19:30 35/3 196.4/182.0 1.07 ± 0.14 1.02 ± 0.28 1.00 ± 0.20 0.99 ± 0.18 0.97 ± 0.18 0.96 ± 0.18 20/17 Mar 1990 16:30–19:30 30/3 223.9/182.0 0.88 ± 0.09 1.08 ± 0.45 1.02 ± 0.34 0.96 ± 0.22 0.90 ± 0.23 0.85 ± 0.23 21/17 Mar 1990 17:20–18:55 76/3 227.6/182.0 0.57 ± 0.11 1.02 ± 0.36 0.94 ± 0.30 0.86 ± 0.27 0.78 ± 0.23 0.72 ± 0.20 22/17 Mar 1990 18:20–20:00 28/3 243.1/182.0 0.42 ± 0.10 0.94 ± 0.35 0.84 ± 0.29 0.75 ± 0.26 0.66 ± 0.21 0.59 ± 0.18 9/7 Apr 1990 16:40–20:00 34/8 146.8/155.0 0.48 ± 0.08 0.86 ± 0.24 0.83 ± 0.21 0.80 ± 0.19 0.76 ± 0.15 0.72 ± 0.12 10/7 Apr 1990 16:00–19:00 124/8 149.3/155.0 0.06 ± 0.01 0.90 ± 0.19 0.79 ± 0.15 0.67 ± 0.12 0.56 ± 0.08 0.45 ± 0.05 11/7 Apr 1990 16:00–20:00 64/8 160.8/155.0 0.57 — 1.01 — 0.99 — 0.97 — 0.94 — 0.91 — 21/14 Mar 1996 18:00–20:30 38/9 70.4/70.8 0.89 — 0.70 — 0.72 — 0.74 — 0.76 — 0.79 — 18/16 Apr 1996 17:00–20:00 25/9 70.1/68.3 1.30 ± 0.24 1.15 ± 0.29 1.08 ± 0.21 1.08 ± 0.17 1.12 ± 0.15 1.22 ± 0.15 17/19 Apr 1999 17:00–20:00 47/12 115.7/110.0 0.31 ± 0.07 0.93 ± 0.19 0.85 ± 0.15 0.76 ± 0.12 0.68 ± 0.09 0.60 ± 0.09 6/13 Apr 2000 17:00–20:00 82/6 177.7/164.0 1.14 — 0.65 — 0.61 — 0.59 — 0.57 — 0.56 — 13/17 Apr 2001 15:00–19:00 50/6 137.0/126.0 0.48 ± 0.08 0.78 ± 0.10 0.74 ± 0.09 0.68 ± 0.07 0.60 ± 0.06 0.53 ± 0.05 18/17 Apr 2001 16:30–19:30 50/6 132.0/126.0 0.62 ± 0.08 0.87 ± 0.11 0.82 ± 0.10 0.78 ± 0.09 0.75 ± 0.09 0.73 ±0.08 Summer

Dates UT Ap F10.7 rNmF2 rNe(160) rNe(170) rNe(180) rNe(190) rNe(200) 6/8 May 1986 18:30–22:00 67/7 69.8/69.5 0.75 ± 0.20 0.88 ± 0.49 0.84 ± 0.40 0.77 ± 0.29 0.71 ± 0.21 0.65 ± 0.18 7/8 May 1986 19:00–22:00 12/7 69.9/69.5 0.81 ± 0.11 0.84 ± 0.40 0.85 ± 0.36 0.84 ± 0.26 0.80 ± 0.16 0.78 ± 0.15 6/8 June 1991 19:00–21:00 49/26 241.3/250.7 1.13 ± 0.18 1.18 ± 0.96 1.21 ± 0.92 1.24 ± 0.89 1.26 ± 0.85 1.29 ± 0.78 9/8 June 1991 19:00–21:00 58/26 245.3/250.7 0.55 ± 0.09 0.99 ± 0.53 0.97 ± 0.48 0.94 ± 0.45 0.88 ± 0.40 0.81 ± 0.36 10/8 June 1991 19:00–21:00 119/26 246.2/250.7 0.43 ± 0.11 0.99 ± 0.53 0.95 ± 0.48 0.89 ± 0.43 0.83 ± 0.38 0.76 ± 0.32 4/1 Aug 1992 17:00–20:00 15/8 130.9/110.3 1.34 ± 0.10 1.14 ± 0.37 1.09 ± 0.10 1.04 ± 0.11 1.02 ± 0.14 1.02 ± 0.13 5/1 Aug 1992 17:30–20:30 35/8 130.5/110.3 0.58 ± 0.24 0.85 ± 0.49 0.87 ± 0.45 0.86 ± 0.44 0.82 ± 0.41 0.76 ± 0.39 14/15 Aug 1994 15:00–18:00 26/13 88.9/81.4 0.81 ± 0.08 0.95 ± 0.06 0.91 ± 0.07 0.89 ± 0.06 0.86 ± 0.06 0.84 ± 0.06 15/6 Jul 2000 15:20–19:20 164/5 213.1/174.3 1.38 ± 0.32 1.05 ± 0.14 1.09 ± 0.16 1.12 ± 0.14 1.12 ± 0.13 1.11 ± 0.12 16/6 Jul 2000 15:30–19:30 50/5 218.9/174.3 G-cond 1.02 ± 0.14 1.05 ± 0.15 1.03 ± 0.14 0.97 ± 0.12 0.89 ± 0.11 Autumn

Dates UT Ap F10.7 rNmF2 rNe(160) rNe(170) rNe(180) rNe(190) rNe(200) 28/25 Sep 1995 15:00–18:00 15/4 72.7/74.0 0.59 — 0.94 — 0.91 — 0.86 — 0.80 — 0.76 — 7/5 Oct 1997 17:00–20:00 44/7 83.5/84.4 0.52 ± 0.06 0.84 ± 0.34 0.77 ± 0.28 0.70 ± 0.21 0.64 ± 0.17 0.59 ± 0.13 19/20 Oct 1998 16:00–20:00 62/22 117.7/121.2 0.35 ± 0.05 0.78 ± 0.40 0.68 ± 0.27 0.59 ± 0.19 0.51 ± 0.14 0.45 ± 0.11 15/7 Oct 1999 17:00–18:30 31/5 198.2/129.4 1.63 ± 0.13 1.05 ± 0.18 1.04 ± 0.14 1.06 ± 0.14 1.08 ± 0.13 1.10 ± 0.10 586 A. V. Mikhailov and K. Schlegel: Geomagnetic storm effects at F1-layer heights

Table 2. List of EISCAT observations (disturbed/reference days), together with daily Ap and F10.7 indices. Observed rNmF2 and rNe at 160–200 km heights are given, together with absolute deviations (dashes correspond to large scatter in data when calculated standard deviations are unreliable). G-conditions means the absence of the F2-layer peak, LT=UT+1

Spring

Dates UT Ap F10.7 rNmF2 rNe(160) rNe(170) rNe(180) rNe(190) rNe(200) 10/9 Apr 1990 13:15–14:15 124/34 149.3/146.8 G-cond 1.12 ± 0.27 1.03 ± 0.29 0.94 ± 0.23 0.84 ± 0.25 0.78 ± 0.24 3/2 Apr 1992 14:00–15:00 32/6 159.7/161.2 0.16 ± 0.07 0.88 ± 0.19 0.83 ± 0.10 0.75 ± 0.12 0.61 ± 0.09 0.47 ± 0.06 20/19 Mar 1996 11:00–12:30 23/15 69.3/69.9 0.63 ± 0.06 0.82 ± 0.27 0.84 ± 0.26 0.83 ± 0.35 0.76 ± 0.21 0.66 ± 0.15 21/19 Mar 1996 11:00–12:30 38/15 70.4/69.9 0.88 ± 0.08 0.87 ± 0.22 0.88 ± 0.23 0.91 ± 0.29 0.91 ± 0.21 0.89 ± 0.25 25/24 Mar 1998 10:30–12:00 16/7 115.0/120.6 0.81 ± 0.15 1.00 ± 0.47 0.95 ± 0.39 0.90 ± 0.46 0.84 ± 0.31 0.78 ± 0.31 26/24 Mar 1998 10:30–12:00 12/7 110.0/120.6 0.73 ± 0.07 1.04 ± 0.42 1.00 ± 0.42 0.95 ± 0.40 0.88 ± 0.29 0.80 ± 0.24 27/24 Mar 1998 10:30–12:00 15/7 108.1/120.6 0.73 ± 0.07 1.04 ± 0.40 1.00 ± 0.44 0.94 ± 0.39 0.84 ± 0.33 0.71 ± 0.28 9/8 Mar 1999 13:00–14:30 21/12 127.1/126.9 0.41 ± 0.06 1.23 ± 0.80 1.04 ± 0.46 0.83 ± 0.58 0.64 ± 0.42 0.49 ± 0.41 10/8 Mar 1999 11:30–13:00 34/12 135.4/126.9 0.63 ± 0.05 1.00 ± 0.55 0.93 ± 0.37 0.82 ± 0.33 0.72 ± 0.31 0.62 ± 0.17 Summer

Dates UT Ap F10.7 rNmF2 rNe(160) rNe(170) rNe(180) rNe(190) rNe(200) 13/6 Aug 1985 14:30–15:30 41/3 68.9/77.9 0.68 ± 0.15 0.89 ± 0.16 0.83 ± 0.12 0.80 ± 0.09 0.76 ± 0.07 0.74 ± 0.06 14/6 Aug 1985 11:30–13:00 11/3 69.3/77.9 0.79 ± 0.03 0.83 ± 0.02 0.82 ± 0.02 0.82 ± 0.02 0.81 ± 0.03 0.81 ± 0.03 14/13 June 1988 12:15–13:30 20/6 111.7/115.0 0.56 ± 0.03 0.82 ± 0.24 0.80 ± 0.39 0.77 ± 0.34 0.72 ± 0.36 0.65 ± 0.23 1/6 Sep 1988 11:00–12:30 21/3 191.3/152.4 0.55 ± 0.03 0.92 ± 0.19 0.88 ± 0.19 0.80 ± 0.21 0.72 ± 0.17 0.65 ± 0.16 28/31 Aug 1989 12:00–13:30 22/6 174.1/208.9 0.83 ± 0.06 0.88 ± 0.38 0.88 ± 0.42 0.89 ± 0.47 0.91 ± 0.40 0.92 ± 0.36 30/31 Aug 1989 10:30–12:00 17/6 192.0/208.9 1.28 ± 0.12 1.05 ± 0.39 1.07 ± 0.55 1.11 ± 0.51 1.14 ± 0.59 1.14 ± 0.81 13/5 Jun 1990 12:20–13:40 70/5 208.8/153.7 0.46 ± 0.03 0.95 ± 0.17 0.90 ± 0.12 0.87 ± 0.15 0.81 ± 0.15 0.74 ± 0.19 5/4 Aug 1992 11:00–12:00 35/15 130.5/130.9 0.52 ± 0.06 0.86 ± 0.05 0.84 ± 0.07 0.80 ± 0.07 0.72 ± 0.08 0.63 ± 0.06 15/14 May 1997 15:00–16:30 56/6 73.0/73.5 G-cond. 1.23 ± 0.71 1.07 ± 0.59 0.91 ± 0.48 0.78 ± 0.37 0.68 ± 0.30 27/26 Jun 1997 10:30–12:00 17/4 71.7/71.8 1.28 ± 0.16 0.91 ± 0.17 0.94 ± 0.18 0.94 ± 0.21 0.97 ± 0.23 0.94 ± 0.24 Autumn

Dates UT Ap F10.7 rNmF2 rNe(160) rNe(170) rNe(180) rNe(190) rNe(200) 22/21 Sep 1987 11:00–13:00 29/10 81.3/83.0 0.83 ± 0.08 0.78 ± 0.23 0.79 ± 0.27 0.80 ± 0.24 0.79 ± 0.27 0.76 ± 0.24 23/21 Sep 1987 11:00–13:00 17/10 80.4/83.0 0.69 ± 0.12 0.88 ± 0.26 0.87 ± 0.31 0.85 ± 0.33 0.83 ± 0.31 0.79 ± 0.28 24/23 Sep 1998 10:00–12:00 28/11 135.4/143.2 0.76 ± 0.07 1.04 ± 0.24 1.04 ± 0.33 1.05 ± 0.38 1.04 ± 0.32 1.01 ± 0.36 10/9 Oct 1990 11:30–12:30 48/12 194.7/183.9 0.28 ± 0.03 0.74 ± 0.21 0.67 ± 0.22 0.59 ± 0.21 0.49 ± 0.16 0.39 ± 0.11 11/9 Oct 1990 10:00–11:00 42/12 205.1/183.9 0.67 ± 0.06 0.73 ± 0.18 0.67 ± 0.22 0.59 ± 0.26 0.49 ± 0.16 0.40 ± 0.12 24/25 Oct 1990 11:00–12:30 25/9 157.5/161.8 1.24 ± 0.05 0.94 — 0.95 — 0.92 — 0.85 — 0.76 — 24/25 Oct 1990 13:30–15:00 25/9 157.5/161.8 0.53 ± 0.14 0.88 — 0.79 — 0.72 — 0.64 — 0.56 —

only slightly disturbed (e.g. 9 April 1990, 4 August 1992 heights (160 km), and this may be due to particle precipita- from Table 2). The choice of 8 June 1991 and 20 October tion effects during disturbed periods. 1998 (Table 1) as the reference days is due to the absence of Usually the amplitude of deviation increases with height available quiet days nearby. (a decrease in rNe in Tables 1 and 2), especially for negative storm effect, but the inverse type of dependence is possible as well (26/23 January 1993, 11/14 January 1990; 14/15 Febru- 3 Data analysis ary 2001; 4/1 August 1992 from Table 1 or 21/19 March Millstone Hill observations (Table 1) show both negative and 1996, 28/31 August 1989 from Table 2). positive storm effects for all seasons, but negative deviations For the convenience of presentation we will consider prevail. As a rule, positive deviations are relatively small δNe = rNe − 1, together with rNe. Figure 1 compared to negative ones. At EISCAT (the auroral zone) shows the δNe (at 180 km) versus δNmF2 = (NmF2dist – negative deviations dominate (Table 2). Some cases of small NmF2ref)/NmF2ref dependence for Millstone Hill and EIS- positive storm effects (30/31 August 1989, 24/23 September CAT. Although the majority of points are clustering in the 1998) may be related to uncertainties of foF2 readings from negative sector of the plots, there is an obvious relationship ionograms used for the Ne(h) proﬁle normalization. On the between the analyzed parameters. To check whether this de- other hand, positive deviations are more pronounced at lower pendence is signiﬁcant, we put together the data for all sea- A. V. Mikhailov and K. Schlegel: Geomagnetic storm effects at F1-layer heights 587

Table 3. Correlation between δNe at F1-layer heights and δNmF2 3. There is a direct and signiﬁcant correlation between for Millstone Hill and EISCAT δNe at F1-layer heights and δNmF2, the correlation co- efﬁcients increasing with height.

Millstone Hill EISCAT 4. At Millstone Hill the summer storm effects in δNe are Height, Corr. F param. Confid. Corr. F param. Confid. less compared to other seasons, but they are well de- km coeff. level, % coeff. level, % tectable, nevertheless. At EISCAT an autumn/spring 160 0.38 5.80 95 0.14 0.45 None asymmetry in the storm effect is well pronounced, while 170 0.50 11.7 99 0.44 5.18 95 the summer effect takes place only with respect to the 180 0.63 23.4 99 0.65 16.3 99 autumnal season. These differences are statistically sig- 190 0.73 40.0 99 0.75 28.7 99 nificant. 200 0.80 62.0 99 0.77 31.2 99 5. It should be stressed that in general a relatively small sensitivity of the F1-region to geomagnetic disturbances exists despite large storm perturbations in the thermo- sons from Tables 1 and 2 and calculated the correlation co- spheric parameters (see later). This is quite different efficients between δNe (180) and δNmF2. The test of sta- from the F2-layer storm behavior. tistical significance for this correlation was made with the Fisher’s F-criterion (Pollard, 1977). The results are given in An interpretation of the revealed morphological features is Table 3. The correlation coefficients are seen to be not very given below based on the contemporary understanding of the large but they are significant at the 95–99% confidence level. ionosphere formation at F1-layer heights. The observations at 160 km show large scatter and look less reliable compared to higher altitudes. The results by Buresova and Lastovicka (2001) and 4 Model calculations Buresova et al. (2002) show some seasonal effect (winter/summer and autumn/spring asymmetry) in the NeF1 re- To understand the physical mechanism of the revealed F1- action to the geomagnetic disturbances. We have calculated layer storm effects, one should consider the aeronomic pa- average rNe from Tables 1 and 2 for different seasons. These rameters responsible for the F1-layer formation during quiet average values, along with the standard deviations, are given and disturbed conditions. A method proposed by Mikhailov in Table 4 for Millstone Hill and EISCAT. The storm ef- and Schlegel (1997), with later modifications (Mikhailov fect is seen to be less in summer compared to other seasons and Forster,¨ 1999; Mikhailov and Schlegel, 2000) applied at Millstone Hill. At EISCAT this summer effect is seen earlier to Millstone Hill observations (Mikhailov and Fos- only with respect to autumn, but an autumn/spring asymme- ter, 1997; Mikhailov and Forster,¨ 1997, 1999) and EIS- try is clearly present in the storm effect. To check whether CAT observations (Mikhailov and Schlegel, 1998; Mikhailov these seasonal differences are significant, we put together and Kofman, 2001) is used here. It allows us to find in a all heights in Table 4 and applied the Student’s T-criterion self-consistent way neutral composition (O, O2,N2 concen- (Pollard, 1977), which examines whether the difference be- trations), neutral temperature specified by three parameters tween two average values is significant. At Millstone Hill (Tex, T120, S), total EUV solar flux with λ < 1050 A,˚ and the summer/winter difference is significant at the 90%, sum- ion composition. Vertical plasma drift related to thermo- mer/spring – at 99.9%, summer/autumn – at 99.9% confi- spheric winds and electric fields can also be derived with dence level, while the autumn/spring difference is insignif- this method, but they are not used in the present analysis. icant. At EISCAT the summer/autumn difference is signifi- The details of the method may be found in the above refer- cant at the 99%, autumn/spring – at 97.5% confidence level, ences; therefore, only the main idea is sketched here. The + while the summer/spring difference is insignificant. Table 4 model used includes: transport process for O (4S) and pho- + 2 + 2 + 2 Q also shows the decrease with height in calculated rNe, in- tochemical processes only for O ( D),O ( P),O2 (X ), + + + dicating the increase in the storm effect with height for all N , N2 and NO ions in the 120–550 km height range. De- seasons. pending on conditions the height interval can be changed (for The results of the morphological analysis may be summa- instance, to avoid the precipitation effect at lower heights) rized as follows: and the observed electron concentration is used as the bound- ary condition. A two-component model of the solar EUV 1. The storm effects may be positive and negative but neg- from Nusinov (1992) is used to calculate the photoionization ative deviations prevail. As a rule, positive storm effects rates in 35-wavelength intervals (100–1050 A).˚ The photo- are small compared to negative ones. ionization and photo-absorption cross sections are obtained from Torr et al. (1979) and Richards and Torr (1988). Flow- + 2. Usually the amplitude of the storm effect increases with ing afterglow laboratory measurements of the O + N2 re- height (especially for negative ones), but cases with the action rate constant by Hierl et al. (1997) are included as inverse type of dependence also take place both at Mill- well. Vertical plasma drift used in the continuity equation stone Hill and EISCAT. for O+ ions is obtained from the observed parameters as a 588 A. V. Mikhailov and K. Schlegel: Geomagnetic storm effects at F1-layer heights

Table 4. Observed seasonal variation of the rNe values at F1-layer heights for Millstone Hill and EISCAT. Average daytime values along with standard deviations are given

Height Millstone Hill EISCAT km Winter Spring Summer Autumn Spring Summer Autumn 160 0.99 ± 0.27 0.92 ± 0.15 0.99 ± 0.12 0.90 ± 0.12 1.00 ± 0.13 0.93 ± 0.12 0.86 ± 0.11 170 0.92 ± 0.24 0.86 ± 0.14 0.98 ± 0.12 0.85 ± 0.16 0.94 ± 0.08 0.90 ± 0.10 0.83 ± 0.14 180 0.87 ± 0.23 0.82 ± 0.14 0.96 ± 0.14 0.80 ± 0.20 0.87 ± 0.07 0.87 ± 0.10 0.79 ± 0.17 190 0.83 ± 0.23 0.77 ± 0.17 0.93 ± 0.17 0.76 ± 0.24 0.78 ± 0.11 0.83 ± 0.13 0.73 ± 0.20 200 0.81 ± 0.25 0.74 ± 0.21 0.89 ± 0.19 0.72 ± 0.28 0.69 ± 0.14 0.79 ± 0.16 0.67 ± 0.23

difference between measured total vertical plasma velocity and diffusion velocity for O+ ions. Collisions of O+ ions + + + + with neutral O, O2,N2 and NO ,O2 ,N2 ,N ions are taken into account. All O+ ion collision frequencies are taken from Banks and Kockarts (1973). Ion concentrations are known at each iteration step from fitting calculated Ne(h) to the experimental ones. Using standard multi-regressional methods we fit the calculated Ne(h) profile to the observed one and find the earlier mentioned aeronomic parameters. The estimated accuracy of the extracted thermospheric parameters is about ±20–25% (Mikhailov and Schlegel, 1997, 2000). Four cases (disturbed/quiet) of days presenting different seasons were analyzed. Winter and spring are presented by Millstone Hill observations on 7/9 November 1997 and 22/17 March 1990, with a well pronounced negative storm effect and increasing δNe with height. Summer and autumn are presented by EISCAT observations on 13/5 June 1990 and 1/6 September 1988, with small storm effects at F1-layer heights. In fact, the 1/6 September case may be prescribed to summer and later, we will refer to June 1990 and Septem- ber 1988 periods as “summer”. Since winter and equinoctial periods demonstrate similar storm effects (Table 4), we will refer to November 1997 and March 1990 periods as “winter” in further discussion. Millstone Hill observations during a very severe geomagnetic storm on 15–16 July 2000 were also analyzed. This period considered by Buresova et al. (2002) is distinguished by small F1-layer storm effects, despite the extremely strong geomagnetic storm with Ap index up to 164. The selected EISCAT observations are characterized by small (E < 7 mV/m) electric fields in order not to intro- duce additional effects (St.-Maurice and Schunk, 1979; Hu- bert and Kinzelin, 1992) to the model calculations. This allows us to consider the daytime auroral ionosphere as a mid- latitude one (Farmer et al., 1984; Lathuilliere` and Brekke, 1985). However, EISCAT is located in the auroral zone, and particle precipitation takes place in some cases during disturbed periods. For instance, 25 September 1998 was not Fig. 2. Observed (corrected on ion composition) and calculated used in our analysis for this reason; a pronounced increase Ne(h) profiles for winter (7/9 November 1997), summer (13/5 June in Ne obviously related with particle precipitation took place 1990) and a very severe geomagnetic storm on 15–16 July 2000. at lower heights on 9 March 1999. But such cases are not numerous in our analysis. The experimental Ne(h), T e(h), T i(h) profiles were cor- tion (O+/Ne ratio) was strongly different from the model rected for disturbed days when the calculated ion composi- one used in the incoherent scatter data analysis (Waldteufel, A. V. Mikhailov and K. Schlegel: Geomagnetic storm effects at F1-layer heights 589

+ Fig. 3. Calculated distribution of atomic oxygen O and molecular Fig. 4. Same as Fig. 3, but a simple analytical approach (Eq. 4) was + M ions as well as their sum, Ne for winter (7 Nov/9 Nov 1997) used in the calculations. Note a close resemblance with Fig. 3. and summer (13 Jun/5 Jun 1990) disturbances. Solid lines and filled symbols refer to the disturbed days. followed by the dissociative recombination of molecular ions with electrons. This scheme of processes may be written as 1971; Mikhailov and Schlegel, 1997). q(O+) = [O+]{γ [N ] + γ [O ]} The observed and calculated Ne(h) profiles for winter 1 2 2 2 + = [ ][ +] 7/9 November 1997, summer 5/13 June 1990, and the se- q(N2 ) γ3 O N2 vere summer storm conditions of 15–16 July 2000 are shown + + [ ][ +] = [ +] q(O2 ) γ2 O2 O α2 O2 Ne in Fig. 2. The selected periods are seen to exhibit strong + γ [N ][O+] + γ [O][N ] = α [NO+]N storm effects in the F2-region, while the F1-region effects are 1 2 3 2 1 e = [ +] + [ +] + [ +] small, especially in summer. The 15–16 July 2000 period is Ne O O2 NO , (1) very interesting, demonstrating a positive storm effect on 15 where qi – primary ion production rates, γi – ion-molecule July and so-called “G-conditions” on 16 July, when the F2- reaction rate coefficients, and αi – dissociative recombina- layer maximum has completely disappeared at usual heights. + tion rate coefficients. Equilibrium concentration of N2 ions The quality of model Ne(h) fitting may be considered as is negligible compared to the main ions (e.g. Goldberg and acceptable. For further analysis we will need ion composi- + Blumle, 1970). tion. The calculated distribution of O and molecular ions For the sake of simplicity we consider in accordance with + = + + + M O2 NO concentrations are shown in Fig. 3 for Ivanov-Kholodny and Nikoljsky (1969) the ionosphere at quiet (QD) and disturbed (DD) days. F1-region heights consisting of atomic O+ and molecular + = + + + + M NO O2 ions. From Eq. (1) we have for O ions + + q(O ) = β[O ], where β = γ1[N2]+γ2[O2] and for molec- 5 Interpretation + = + + + ular ions M NO O2 , both produced by direct pho- Physical interpretation of the obtained results may be given toionization and via ion-molecule reactions, we may write + + + from an analysis of a simple scheme of photochemical q(O ) + q(M ) = q = αave[M ]Ne, (2) processes controlling the daytime ionosphere at F1-layer heights. The main processes are photoionization of neutral where + [ +] [O], [O2], [N2] by solar EUV radiation, the conversion of [NO ] O2 αave = α + α primary ions to molecular ones via ion-molecule reactions, 1 [M+] 2 [M+] 590 A. V. Mikhailov and K. Schlegel: Geomagnetic storm effects at F1-layer heights

layer storm effects in winter and equinoxes compared to summer. The differences (disturbed minus quiet) for [O+] and [M+], as well as for Ne at 180 km, are shown in Table 5 for “winter” and “summer” cases. Table 5 shows that [O+] and [M+] change in opposite directions. Normally [O+] decreases while [M+] always increases in disturbed conditions. The [O+] decrease usually prevails, especially in “winter”, and this determines the negative sign of the storm effect observed in the majority of cases (Tables 1 and 2). The “summer” [M+] increase is larger compared to the “winter” one, and the resultant negative storm effect is less in summer due to this compensation. The 15–16 July 2000 storm effects will be discussed later. Let us consider the aeronomic parameters responsible for such seasonal differences in [O+] and [M+] storm-time variations. Table 6 gives calculated changes (at 180 km) in [O], + [O2], [N2], q(O ), in total ion production rate q, as well as in linear loss coefficient β. Generally, the atomic oxygen ion concentration decreases for disturbed periods, and this is due to a decrease in q(O+) and an increase in β (see the first term in Eq. 4). The molecular ion concentration [M+] always increases for the storm periods (Table 5), but this is due to different reasons in “summer” and in “winter”. The second term of Eq. (4) indicates that [M+] depends on three parameters. Our analysis has shown that the increase of the total ion production rate q is the main channel for the [M+] increase in “summer”, while αave and Ne changes are less, and they work in opposite directions, compensating each other to a great extent. In “winter” q variations are small (Fig. 5 and Fig. 5. Calculated ion production rates for the analyzed distur- Table 6) and the main contribution to the [M+] increase pro- bances. Again solid lines refer to disturbed days. vides the decrease in Ne, as this leads to a decrease in the M+ recombination rate. Figure 5 (top panel) shows that quiet + + is the average-weighted dissociative recombination rate coef- time q(O ) and q(N2 ) are close at F1-layer heights in “win- ficient. Keeping in mind that [M+] = Ne − [O+] and [O+] ter”, while the disturbed values varying in opposite directions = q(O+)/β, we obtain from Eq. (2) compensate each other to a great extent. In “summer” the + + initial quiet time q(O ) and q(N2 ) values differ essentially q(O+) q + N 2 − N − = 0. (3) (Fig. 5, bottom) and the additional large increase in q(N2 ) e e + + β αave and also in q(O2 ) overpowers the decrease in q(O ). This Equation (3) may be rewritten as results in a noticeable increase in the total ion production rate. + q(O ) q The F1-layer is formed in the vicinity of the total ion pro- Ne = + . (4) β αaveNe duction rate maximum (Fig. 5), so the absorption of EUV radiation is not very large (the optical depth τ = 1) and the The first term in Eq. (4) presents the O+ ion concentration specific q are roughly proportional to the concentration of and the second term – the concentration of M+ ions. The two i the ionized neutral species. Therefore, the calculated varia- terms of Eq. (4), as well as their sum, are shown in Fig. 4 for tions in q roughly reflect the variations in neutral composi- 7/9 November 1997 and 5/13 June 1990 to be compared with i tion. The calculated variations of molecular species concen- the model calculated distributions (Fig. 3) when the complete trations (Table 6) clearly indicate the seasonal difference in set of pertinent processes is taken into account. The two fig- the thermosphere reaction to the geomagnetic disturbances. ures are seen to be very similar; therefore, our simplified ap- This results in larger 1q in “summer”. proach (Eq. 4) may be used for physical interpretation. Some autumn/spring asymmetry in the F1-layer storm 5.1 Seasonal difference in the F1-layer reaction to geomag- effect mentioned earlier can be related to an asymmetry netic storms in atomic oxygen annual variations in the thermosphere, with [O] being more abundant in late autumn compared Our morphological analysis (Table 4), as well as the results to spring, according to the thermospheric model MSIS-86 by Buresova et al. (2002), indicate more pronounced F1- (Hedin, 1987). In general, seasonal variations of the F1- A. V. Mikhailov and K. Schlegel: Geomagnetic storm effects at F1-layer heights 591

Table 5. Differences at 180 km (disturbed-quiet) in atomic and molecular ion concentrations as well as in Ne for “winter” and “summer” disturbances

Dates 1[O+] × 105 cm−3 1[M+] × 105 cm−3 1[Ne] × 105 cm−3 “Winter” 7/9 Nov 1997 −0.95 +0.22 −0.73 22/17 Mar 1990 −1.07 +0.26 −0.81 “Summer” 13/5 June 1990 −0.83 +0.57 −0.26 1/6 Sep 1988 −0.78 +0.41 −0.37 “The 15/16 July 2000 storm” 15/6 Jul 2000 +0.08 +0.14 +0.22 16/6 Jul 2000 −0.22 +0.11 −0.11

Table 6. Calculated variations at 180 km in neutral composition, ion production rates and linear loss coefﬁcient for “winter” and “summer” disturbances

+ Dates 1 log[O] 1 log[O2] 1 log[N2] 1 log q(O ) 1 log q 1 log β “Winter” 7/9 Nov 1997 −0.190 +0.220 +0.170 −0.237 −0.017 +0.259 22/17 Mar 1990 −0.025 +0.245 +0.155 −0.133 −0.022 +0.167 “Summer” 13/5 Jun 1990 −0.084 +0.428 +0.249 −0.069 +0.144 +0.351 1/6 Sep 1988 −0.162 +0.339 +0.256 −0.149 +0.127 +0.304 “The 15/16 July 2000 storm” 15/6 Jul 2000 +0.169 +0.106 +0.05 +0.155 +0.06 +0.07 16/6 July 2000 −0.349 +0.213 −0.02 −0.191 +0.02 +0.08 “Height decreasing positive storm effect” 11/14 Jan 1990 −0.231 −0.075 −0.188 0.00 0.026 −0.134

layer storm effect magnitude reflect seasonal variations in the For the sake of simplicity we may suppose that the ther- atomic oxygen concentration. Large [O] concentration dur- mosphere is isothermal and neutral species [O] and [M] ≈ ing equinoctial and winter seasons results in a large contribu- [N2] follow the barometric law: [O] = [0]exp(−z/H ) and + tion of [O ] ions to Ne. Therefore, any storm-time decrease [M] = [M]0exp(−1.75z/H ), where H = kT n/mg is the in [O] is essential and noticeable in the Ne variations, unlike atomic oxygen scale height. The O+ production rate may be + + in summer when the role of [O ] is much less compared to written as q(O ) = jo[O]exp(−aChχ ), where j is the ion- the influence of [M+]. ization efficiency depending on the incident solar EUV flux In summary, we may conclude that the less pronounced and ionization cross sections, Chχ is the Chapman function “summer” storm effect compared to the “winter” one is due for solar zenith angle χ, and a includes the column density of to a stronger compensation of negative 1[O+] by positive neutrals multiplied by absorption cross sections. The linear 1[M+] in “summer”. In turn, these seasonal differences just loss coefficient β may be written as β = γ [M]. In this case + + reflect the seasonal differences in neutral composition both [O ] = q(O )/β ∝ exp(0.75z/H ). This is an increasing + during quiet and disturbed conditions. function with height. The M -ion concentration following the total ion production rate (Fig. 5) exhibits small height 5.2 Height dependence of the storm effect variations in the F1-region passing via the extreme (Figs. 3 and 4). Therefore, [O+] begins to dominate over [M+] in Normally, the F1-layer storm effect increases with height the upper part of the F1-region and above in the F2-region. [ +] (Tables 1, 2, and 4). This has a simple explanation based Usually, 1 O controls the sign of the F1-layer storm effect on the competition mechanism between 1[O+] and 1[M+]. (Table 5), and this explains the direct and significant correla- 592 A. V. Mikhailov and K. Schlegel: Geomagnetic storm effects at F1-layer heights

Fig. 7. Calculated O+/Ne ratio for the 15–16 July 2000 storm compared to the Millstone Hill model. Note that the quiet time ion composition also strongly differs from the model one.

tions related to this period are, why is the F1-layer storm effect relatively small on 16 July, and what is the reason for the positive effect on 15 July? Our method was applied to the two disturbed days, 15–16 July, and to a quiet reference day on 6 July. The observations were provided by Millstone Hill IS facility. Figure 6 shows the calculated [O+], [M+] distributions and their sum for quiet and disturbed days. The first surpris- Fig. 6. Same as Fig. 3, but for a very severe geomagnetic storm on 15–16 July 2000. ing result concerns the reference day. Although 6 July 2000 and the eight previous days were magnetically quiet, the calculated ion composition (O+/Ne ratio) differs strongly from tion between δNe and δNmF2 (Table 3). the model one (Oliver, 1975) used at Millstone Hill for in- A smaller sensitivity of the F1-region to geomagnetic dis- coherent scatter data analysis (Fig. 7, dashes). This model presents the quiet-time ion composition which is indepen- turbances compared to the F2-layer usually mentioned in lit- + erature is also explained in the framework of this mechanism. dent of geophysical conditions, while the calculated O /Ne In “winter” there is a partial compensation and in “summer” ratio for 6 July corresponds to a disturbed ionosphere en- a practically complete compensation of negative 1[O+] by riched strongly with molecular ions. The peculiarity of 6 July positive 1[M+], while the pure 1[O+] effect takes place in also confirms a comparison with the monthly median IRI- the F2-region. 90 model (Bilitza, 1990) which corresponds to a quiet ionosphere. The IRI-90 gives NmF2 = 6×105 cm−3 for the condi- − 5.3 The 15–16 July 2000 severe storm effect tions in question, while the observed value is 4 × 105 cm 3. A comparison of Fig. 6 with Fig. 3 for the 13/5 June 1990 This was a very severe geomagnetic storm with a well- period shows a strong depletion in [O+] on 6 July 2000, sim- pronounced onset on the afternoon of 15 July and a main ilar to the disturbed day of 13 June 1990. All this indicates phase with Dst decreasing down to −300 nT at 21:00 UT. that the reference day of 6 July 2000 is not a typical sum- But it should be noted that the geomagnetic field was also mer one. It may be attributed to so-called “low” summer disturbed for the two previous days with Ap = 42 on 13 July dates (Ivanov-Kholodny et al., 1981) or to a quiet-time F2- and Ap = 51 on 14 July. A strong positive storm effect took layer deviation discussed by Mikhailov and Schlegel (2001). place in the F2-region on 15 July which extended down to In both cases the effect is related to a relatively large ther- F1-layer heights (Table 1 and Fig. 2, bottom). On the next mospheric neutral composition variation during geomagnet- day, 16 July there was a complete disappearance of the F2- ically quiet periods. Therefore, rather small F1-layer storm layer at usual heights and a maximum in the Ne(h) profile effects on 15–16 July may be related to the peculiarity of the in the F1-layer – the so-called G-condition. Despite such preceding reference period. But, nevertheless, we are consid- severe F2-layer storm effects, implying very strong pertur- ering this interesting period which is discussed in literature bations in the thermospheric parameters, rather small storm (Buresova et al., 2002; Pavlov and Foster, 2001; Basu et al., effects were registered at F1-layer heights (Table 1). This ef- 2001). fect was also mentioned by Buresova et al. (2002) on the Eu- Let us consider the calculated variations of aeronomic pa- ropean ionosondes data analysis. The most interesting ques- rameters for the dates in question (Tables 5 and 6 and Fig. 6). A. V. Mikhailov and K. Schlegel: Geomagnetic storm effects at F1-layer heights 593

The earlier discussed compensation of negative 1[O+] by positive 1[M+] takes place for 16 July (Table 5), while a summation of positive 1[O+] and 1[M+] results in a positive storm effect on 15 July (see later). Both effects are due to the neutral composition variations (Table 6) and correspond- ing changes in q(O+), β and total q. Although the atomic oxygen decrease on 16 July is essential (by 2.2 times), this gives rise to a small effect in 1Ne (Fig. 6). As mentioned earlier, the initial [O+] contribution to Ne is relatively small compared to the [M+] one; therefore, an additional decrease in [O] has practically no effect at F1-layer heights where molecular ions dominate. But this turns out to be crucial for the F2-region, where the normal F2-layer has disappeared (G-condition). This effect was analyzed earlier for Millstone Hill (Mikhailov and Foster, 1997) and EISCAT (Mikhailov and Schlegel, 1998) observations.

5.4 Positive storm effects

Our morphological analysis (Tables 1 and 2) has shown some cases of positive F1-layer storm effect with the magnitude both increasing and decreasing with height. Usually, the positive effects are not large but they are considered as well, in order to make the picture complete. The 15 July 2000 case with the height increasing effect (Figs. 2 and 6) is mostly due to the atomic oxygen increase (Table 6). This increase in [O], along with the enhanced vertical plasma drift (due to Fig. 8. Same as Fig. 3, but for a positive disturbance on 11 Jan/14 the storm-induced equatorward thermospheric wind), results Jan 1990 with a height decreasing storm effect (top) and calculated in a strong positive effect at the F2-layer heights (Fig. 2). So, ion production rates for this case (bottom). this case may be attributed to normal F-layer positive disturbances. A positive storm effect decreasing with height is pre- 6 Discussion sented by the 11/14 January 1990 case (Table 1). The observed Ne(h) profiles show a small positive 1Ne effect in Incoherent scatter observations are known to provide the the bottomside which disappears in the vicinity of the F2- most complete and consistent information on ionospheric layer maximum and appears again in the topside. The cal- plasma in the F-region. Observed Ne(h) profiles were used + + culated 1[O ] and 1[M ] variations are shown in Fig. 8 in the paper for the analysis of the morphology of F1-layer + (top). Below 180 km the summation of both positive 1[O ] storm effects. Our self-consistent method was successfully + and 1[M ] results in a noticeable positive effect in Ne. At used for the physical interpretation of the storm effects. + 180 km the resultant 1Ne is totally due to 1[O ], while Nevertheless, some problems should be mentioned in rela- + + above 180 km 1[O ] and 1[M ] act in opposite directions, tion with the use of IS observations. Unlike ground-based + decreasing the resultant 1Ne. Although 1[O ] dominates ionosonde data the IS observations are irregular in time and + over 1[M ] above 180 km, its magnitude decreases with it is not always possible to find a suitable combination of height and there is no storm effect at 220 km. disturbed/quiet days for the analysis; therefore, the usable The calculated variation of aeronomic parameters is given amount of data is limited. The quality of the experimental in Table 6 (bottom line) and the ion production rates are material is different at Millstone Hill and EISCAT. Due to shown in Fig. 8 (bottom). An interesting result is a decrease rare (usually 3 per hour) observations at Millstone Hill, we in concentration of all neutral species for the disturbed day. had to use a 3–4 h period to calculate median profiles. It is Despite this decrease in [O], we have positive 1[O+], which not always possible (especially for disturbed days) to find a is mainly due to the 1β decrease (see Eq. 4). The latter is 3–4 h period around noon of relative stability in NmF2 and clearly seen at 180 km, where 1q(O+) = 0 (Table 6). Pos- hmF2 variations. This criterion is applied to decrease the itive 1[M+] below 180 km is mainly due to the increase in scatter in the observations and increase the reliability of me- the total ion production rate (Fig. 8, bottom and Eq. 4). All dian profiles. Unlike EISCAT observations there is a problem ion production rate profiles are seen to be shifted down by with using the routine V z(h) data at Millstone Hill, as they ≈ 20 km for the disturbed day and this is quite different from may need an additional correction due to technical reasons. the other storm cases considered. Therefore, not all available routine Millstone Hill observa- 594 A. V. Mikhailov and K. Schlegel: Geomagnetic storm effects at F1-layer heights tions can be used for our model calculations. show a seasonal difference in 1(N2/O) at 280 km for middle Another well-known general problem with IS observations latitudes and daytime hours (their Fig. 4). According to the is related to the ion composition, and this is especially im- model of thermospheric composition by Zuzic et al. (1997) portant in case of the F1-layer data analysis. Experimen- based on the ESRO-4 observations, larger N2/O disturbance tal T e(h), T i(h) and Ne(h) profiles derived from the in- intensity corresponds to larger 1[N2] and smaller 1[O], al- coherent scatter data analysis depend on the ion composi- though the latter demonstrates large scatter (their Fig. 6). tion used in the fit of the theoretical to the measured auto- This seasonal difference in the thermospheric reaction may correlation function (ACF). An uncertainty in ion compo- be attributed to the enhanced Joule heating in summer and to sition may lead to considerable uncertainties in the derived the storm-induced/background thermospheric winds interac- T e(h) and T i(h) profiles and to somewhat smaller uncer- tion (e.g. Fuller-Rowell and Codrescu, 1996). tainties in Ne(h) (e.g. Waldteufel, 1971; Lathuillere` et al., 1983). The effect of varying ion composition is most noticeable during disturbed periods, but an appreciable effect Unlike the F2-layer, where the role of dynamical processes may also take place for quiet periods as well (e.g. 6 July (thermospheric winds, in particular) is essential, the elec- 2000). The largest uncertainties take place at the F1-region tron concentration in the F1-region (heights below 200 km) heights, where the ion composition changes from molecular is controlled by photochemical processes. Therefore, the ob- to atomic one. Therefore, a correction of the experimental served NeF 1 variations just reflect the variations of neutral T e(h), T i(h) and Ne(h) profiles is required. A simple cor- composition and temperature. By analogy with the F2-layer rection proposed by Waldteufel (1971) may be applied when (Prolss,¨ 1993), a decrease in the O/(N2 + O2) ratio results the deviations in ion composition from the model (used in the in the negative storm effect, while F1-layer positive storm IS data analysis) are not large, but in the case of strong pertur- effect may be due to some other mechanisms. In particular, bations, such as those present in the July 2000 event (Fig. 7), the positive storm effect at F1-layer heights on 15 July 2000 this simple correction results in unreal T e(h) and T i(h) pro- is mostly due to the atomic oxygen increase, as our analysis files. In such cases a more sophisticated iterative method has shown. The height decreasing positive storm effect on 11 considered by Mikhailov and Schlegel (1997) should be ap- January 1990 (Fig. 8), on the other hand, was shown to be due plied, which provides the proper fit to the measured ACF. to a thermosphere contraction below 220 km which resulted This iterative method requires considerable calculations and in a decrease of neutral concentrations at a given height (e.g. cannot be applied routinely. at 180 km in Table 6). This decrease in neutral atmosphere EISCAT observations (each particular experiment) are not density has clearly appeared in the total ion production rate normalized by foF2 values. Although the declared uncer- (Fig. 8, bottom). The q(h) profile is shifted down as a whole tainty in the measured electron concentration is not large without change in the maximum production rate. Such type −10 ÷ 12% (Farmer et al., 1984; Kirkwood et al., 1986), of neutral atmosphere variation may be related to a passage this may result in wrong relative deviations δNe when dif- of planetary or gravity waves during disturbed periods (e.g. ferent (quiet/disturbed) days are compared. The absence of Burns and Killeen, 1992). such normalization also results in a shift between long-pulse and multi-pulse Ne(h) profiles, and this shift should be taken into account before the profiles are used for analysis. Un- Coming back to the small storm effect on 16 July 2000 fortunately, foF2 data (due to problems with ground-based related to the peculiarity of the reference day 6 July, an addi- ionosonde observations in the auroral zone) are not available tional argument may be mentioned. The calculated total neu- in some interesting cases, and one has to use IS observations, tral density is ρ = 1.01 × 10−11 g cm−3 at 180 km, and this as they are without normalization. Despite the problems en- is twice as much as on 5 June 1990, although the geophysical countered, the IS observations provide necessary information conditions were close for the two days. For the disturbed day to specify physical mechanisms of the F1-layer storm effects, 13 June 1990 the neutral density ρ has increased by a factor and this was implemented using our self-consistent method. of 1.6 while, ρ has turned out to be practically unchanged Among the observed storm features at mid-latitudes, the on 16 July 2000 with respect to the reference day of 6 July most interesting is a seasonal difference with a less pro- 2000. Since neutral density at 180 km is mainly represented nounced summer storm effect compared to other seasons. by molecular nitrogen N2, such a stability in ρ may be an This implies a seasonal difference in the thermosphere re- indication of a “saturation” of the lower thermosphere with action to geomagnetic disturbances with the summer [O2] molecular species. With more or less a fixed concentration and [N2] increase, larger than the winter one (see Table 6 of N2 at the turbopause level and similar temperatures, one [ ] [ ] + for 1 log O2 and 1 log N2 and Fig. 5 for 1q(N2 ) and should have similar [N2] values. Indeed, the neutral tem- + 1q(O2 )). The MSIS-86 model (Hedin, 1987) gives very peratures (calculated at 180 km) are similar – 1158 K on 6 small seasonal differences in 1[O2] and 1[N2], even at the July and 1161 K on 16 July, and this resulted in similar N2 F2-region heights (280 km) and practically negligible 1[O2] and O2 concentrations for the two days. Since Ne reflects and 1[N2] variations in the F1-region (180 km). On the other mainly molecular ions (see Fig. 6, bottom), we have a small hand, the ESRO-4 observations (Prolss¨ and von Zahn, 1977) storm effect at F1-layer heights on 16 July 2000. A. V. Mikhailov and K. Schlegel: Geomagnetic storm effects at F1-layer heights 595

7 Conclusions 8. The concentration of O+ ions increases with height, while [M+] demonstrates small height variations in the Our morphological analysis of the storm effects at F1-layer F1-region. Therefore, [O+] begins to dominate over heights (160–200 km) using EISCAT and Millstone Hill in- [M+] in the upper part of the F1-region and above in the coherent scatter observations followed by physical interpre- F2-region. Usually, 1[O+] controls the sign of the F1- tation has shown the following. layer storm effect, and this explains both the increase of The morphological results: the storm effect with height and the direct (significant) correlation between δNe and δNmF2. 1. The storm effects in Ne may be positive and negative, but negative deviations prevail for all seasons. The mag- 9. Small, positive storm effects also observed in the F1- nitude of the positive effect is smaller compared to the region may result from different reasons. The increase negative one. of atomic oxygen abundance results in a positive effect with a height increasing magnitude. A contraction 2. Usually, the amplitude of deviation increases with of the lower thermosphere (presumably due to gravity height, but the inverse type of dependence may also take waves) results in a height decreasing positive storm ef- place. fect. 3. There is a direct and significant correlation between 10. In summary one may conclude that all the observed δNe at the F1-layer heights and δNmF2, the correlation F1-layer storm effects may be related to seasonal and coefficients increasing with height. storm-time variations of neutral composition (O, O2, N2). The present day understanding of the F1-region 4. At middle latitudes the summer storm effect is less com- formation mechanisms is sufficient to explain the ob- pared to other seasons, but it is well detectable. In the served storm effects. auroral zone this summer effect takes place only with respect to the autumnal period, while a spring/autumn Acknowledgements. The authors thank the Millstone Hill Group of asymmetry in the storm effect is well pronounced. the Massachusetts Institute of Technology, Westford; also the Di- rector and the staff of EISCAT for running the radar and providing 5. The F1-region exhibits a relatively small reaction to the data. The EISCAT Scientific Association is funded by scien- geomagnetic disturbances despite large storm perturba- tific agencies of Finland (SA), France (CNRC), Germany (MPG), tions in the thermospheric parameters at the heights in Japan (NIPR), Norway (NF), Sweden (NFR), and the United King- dom (PPARC). A. V. Mikhailov is grateful to the Max-Planck- question. Gesellschaft for a research stipend at the MPAE. As the F1-region is formed, where both atomic O+ and Topical Editor M. Lester thanks C. Huang and another referee + = + + + for their help in evaluating this paper. molecular M O2 NO ions strongly contribute to electron concentration Ne, the competition between 1[O+] and 1[M+] contributions explains the main morphological References features, in particular: Antonova, L. A. and Ivanov-Kholodny, G. S.: The F1-layer. Condi- 6. Normally [O+] decreases while [M+] always increases tions for appearance and height, Geomagn. Aeronom., 28, 813– 816, 1988a. in disturbed conditions. The [O+] decrease usually pre- Antonova, L. A. and Ivanov-Kholodny, G. S.: The F1-layer. De- vails (especially during equinoxes and in winter) and pendence of electron concentration at the layer maximum on he- this determines the negative sign of the storm effect ob- liogeophysical conditions, Geomagn. Aeronom., 28, 817–819, + served in the majority of cases. The summer [M ] in- 1988b. crease is larger compared to the winter one, and the re- Banks, P. M. and Kockarts, G.: Aeronomy, Academic Press, New sultant negative storm effect is smaller in summer due to York, London, 1973. this compensation. The latter is related to seasonal dif- Basu, S., Basu, Su., Groves, K. M., Yeh, H.-C., et al.: Response ferences in the thermosphere reaction to geomagnetic of the equatorial ionosphere in the South Atlantic region to the great magnetic storm of 15 July 2000, Geophys. Res. Lett., 18, disturbances, with the summer [O2] and [N2] increase larger than the winter one. This revealed seasonal effect 3577–3580, 2001. is confirmed by ESRO-4 neutral composition observa- Bilitza, D. (Ed): International Reference Ionosphere IRI-90, URSI- COSPAR, Rep. NSSDC/WDC-A Rocket and Satellites, Green- tions for disturbed conditions. belt, USA, 1990. 7. The small sensitivity of the F1-region to geomagnetic Buonsanto, M. J.: Ionospheric storms – A review, Space Sci. Rev. 88, 563–601, 1999. disturbances compared to the F2-layer one is also ex- Buresova, D. and Lastovicka, J.: Changes in the F1-region elec- plained in the framework of this mechanism. There is tron density during geomagnetic storms at low solar activity, J. partial compensation (in winter and equinoxes) or prac- Atmos. Solar-Terr. Phys., 63, 537–544, 2001. tically complete (in summer) compensation of negative Buresova, D., Lastovicka, J., Atadill, D., and Miro, G.: Daytime + + 1[O ] by positive 1[M ] at F1-layer heights, while electron density at the F1-region in Europe during geomagnetic the pure 1[O+] effect takes place in the F2-region. storms, Ann. Geophysicae, 20, 1007–1021, 2002. 596 A. V. Mikhailov and K. Schlegel: Geomagnetic storm effects at F1-layer heights

Burns, A. G. and Killeen, T. L.: The equatorial neutral thermo- Mikhailov, A. V. and Schlegel, K.: A self-consistent estimate of + spheric response to geomagnetic forcing, Geophys. Res. Lett., O + N2 rate coefficient and total EUV solar flux with λ1050 A˚ 19, 977–980, 1992. using EISCAT observations. Ann. Geophysicae, 18, 1164–1171, Danilov, A. D. and Lastovicka, J.: Effects of geomagnetic storms 2000. on the ionosphere and atmosphere, Int. J. Geomagn. Aeronom., Mikhailov, A. V. and Schlegel, K.: Equinoctial transitions in the 2, 209–224, 2001. ionosphere and thermosphere, Ann. Geophysicae, 19, 783–796, Farmer, A. D., Lockwood, M., Horne, R. B., Bromage, B. J. I., and 2001. Freeman, K. S. C.: Field-perpendicular and field-aligned plasma Nusinov, A. A.: Models for prediction of EUV and X-ray solar ra- flows observed by EISCAT during a prolonged period of north- diation based on 10.7 cm radio emission., Proc. Workshop on So- ward IMF, J. Atmos. Terr. Phys., 46, 473–488, 1984. lar Electromagnetic Radiation for Solar Cycle 22, Boulder, Co., Fuller-Rowell, T. J. and Codrescu, M. V.: On the seasonal response July 1992, (Ed) Donnely, R. F., NOAA ERL. Boulder, Co., USA, of the thermosphere and ionosphere to geomagnetic storms, J. 354–359, 1992. Geophys. Res., 101, 2343–2353, 1996. Oliver, W. L.: Models of the F1-region ion composition variations, Goldberg, R. A. and Blumle, L. J.: Positive composition from a J. Atmos. Terr. Phys., 37, 1065–1076, 1975. rocket-borne mass-spectrometre, J. Geophys. Res., 75, 133–142, Pavlov, A. V. and Foster, J. C.: Model/data comparison of F-region 1970. ionospheric perturbation over Millstone Hill during the severe Hedin, A. E.: MSIS-86 thermospheric model, J. Geophys. Res., 92, geomagnetic storm of 15–16 July 2000, J. Geophys. Res., 106, 4649–4662, 1987. 29 051–29 069, 2001. Hierl, P. M., Dotan, I., Seeley, J. V., Van Doran, J. M., Morris, R. Pollard, J. H.: A handbook of numerical and statistical techniques, and Viggiano, A. A.: Rate coefficients for the reactions of O+ Camb. Univ. Press, 1977. with N2 and O2 as a function of temperature (300–1800 K), J. Prolss,¨ G. W.: On explaining the local time variation of ionospheric Chem. Phys., 106, 9, 3540–3544, 1997. storm effects, Ann. Geophysicae, 11, 1–9, 1993. Hubert, D. and Kinzelin, E.: Atomic and molecular ion tempera- Prolss,¨ G. W.: Ionospheric F-region storms, in Handbook of Atmo- tures and ion anisotropy in the auroral F-region in the presence spheric Electrodynamics, Vol. 2 (Ed) Volland, CRC Press/Boca of large electric fields, J. Geophys. Res., 97, 1053–1059, 1992. Raton, 195–248, 1995. Ivanov-Kholodny, G. S. and Nikoljsky G. M.: The Sun and the Prolss,¨ G. W. and von Zahn, U.: Seasonal variations in the latitudi- Ionosphere, Nauka, M., 335, in Russian, 1969. nal structure of atmospheric disturbances, J. Geophys. Res., 82, Ivanov-Kholodny, G. S., Mikhailov, A. V., and Ostrovsky, G. I.: 5629–5632, 1977. Day-to-day change in the summer values of hmF2 and NmF2 as Richards, P. G. and Torr, D. G.: Ratios of photoelectron to EUV ion- a reflection of variations in the neutral composition of the upper ization rates for aeronomic studies, J. Geophys. Res., 93, 4060– atmosphere, Geomagn. Aeronom., 21, 615–617, 1981. 4066, 1988. Kirkwood, S., Collis, P. N., and Schmidt, W.: Calibration of elec- Richmond, A. D.: Upper-atmospheric effects of magnetic storms: a tron densities for EISCAT UHF radar, J. Atmos. Terr. Phys., 48, brief tutorial, J. Atmos. Solar-Terr. Phys., 62, 1115–1127, 2000. 773–775, 1986. Shchepkin, L. A.: Some patterns in the vertical distribution of elec- Lathuillere,` C., Lejeune, G., and Kofman, W.: Direct measurements tron and ion concentrations at heights between 130–200 km, Ge- of ion composition with EISCAT in the high-latitude F1-region, omagn. Aeronom., 9, 380–384, 1969. Radio Sci., 18, 887–893, 1983. Shchepkin, L. A., Shuyskaya, Z. I., and Koshelev, V. V.: Estimation Lathuillere,` C. and Brekke, A.: Ion composition in the auroral iono- of the seasonal variation of the gas concentration at the height sphere as observed by EISCAT, Ann. Geophysicae, 3, 557–568, of the F1-layer from data on the conditions of its development, 1985. Geomagn. Aeronom., 12, 870–872, 1972. Mikhailov, A. V.: Ionospheric F2-layer storms, Fisica de la Tierra, Shchepkin, L. A. and Vinitzky, A. V.: Derivation of the foF1 char- 12, 223–262, 2000. acteristic from ionograms, Geomagn. Aeronom., 21, 257–258, Mikhailov, A. V. and Foster, J. C.: Daytime thermosphere above 1981. Millstone Hill during severe geomagnetic storm, J. Geophys. St.-Maurice, J.-P. and Schunk, R. W.: Ion velocity distributions in Res., 102, 17,275–17,282, 1997. the high-latitude ionosphere, Rev. Geophys. Space Phys., 17, 99– Mikhailov, A. V. and Forster,¨ M.: Day-to-day thermosphere param- 134, 1979. eter variation as deduced from Millstone Hill incoherent scatter Torr, M. R., Torr, D. G., Ong, R. A., and Hinteregger, H. E.: Ionoza- radar observations during 16–22 March 1990 magnetic storm pe- tion frequencies for major thermospheric constituents as a func- riod, Ann. Geophysicae, 15, 1429–1438, 1997. tion of solar cycle 21, Geophys. Res. Lett., 6, 771–774, 1979. Mikhailov, A. V. and Forster,¨ M.: Some F2-layer effects during the Vinitzky, A. V., Sukhomazova, G. I., and Shchepkin, L. A.: Rela- 6–11 January 1997 CEDAR storm period as observed with the tionship between the properties of the ionospheric F2- and F1- Millstone Hill incoherent scatter facility, J. Atmos. Solar-Terr. layers in the day-to-day variations, Geomagn. Aeronom., 22, Phys, 61, 249–261, 1999. 707–709, 1982. Mikhailov, A. V. and Kofman, W.: An interpretation of ion com- Waldteufel, P.: Combined incoherent scatter F1-region observa- position diurnal variation deduced from EISCAT observations, tions, J. Geophys. Res., 76, 6995–6999, 1971. Ann. Geophysicae, 19, 351–358, 2001. Zevakina, R. A., Goncharova, E. E., Pushkova, G. N., and Yo- Mikhailov, A. V. and Schlegel, K.: Self-consistent modeling of dovich, L. A.: Effect of the ring current on on ionospheric F- the daytime electron density profile in the ionospheric F-region, region disturbance, Geomagn. Aeronom., 11, 825–829, 1971. Ann. Geophysicae, 15, 314–326, 1997. Zuzic, M., Scherliess, L., and Prolss,¨ G. W.: Latitudinal structure of Mikhailov, A. V. and Schlegel, K.: Physical mechanism of strong thermospheric composition perturbations, J. Atmos. Solar-Terr. negative storm effects in the daytime ionospheric F2-region ob- Phys., 59, 711–724, 1997. served with EISCAT, Ann. Geophysicae, 16, 602–608, 1998.